Template-driven Documentation of Best Practices

(1)

T E M P L A T E - D R I V E N D O C U M E N T A T I O N

O F B E S T P R A C T I C E S

Meshari Alwazae

Report Series / Department of

Computer and Systems Sciences

No. 15-017

(2)

(3)

Template-driven Documentation

of Best Practices

(4)

ISBN 978-91-7649-286-4

Printed in Sweden by Holmbergs, Malmö 2015

(5)

Dedicated to my parents,

wife, daughter and son as

well as to the memory of

my late brother, Mansor,

who passed away during

the second year of my PhD

(6)

Abstract

Knowledge Management has become a key instrument for identifying, creat-ing and sharcreat-ing organizational knowledge assets. An attractive means for sharing knowledge is Best Practices (BPs), which are proven as well as effi-cient and effective solutions to recurring problems. BPs can offer significant benefits, including improved performance, reduced re-work and cost sav-ings. However, it is challenging to share and use BP Documents (BPDs) in organizations, which is the overall problem addressed in this thesis. More precisely, the thesis focuses on two sub-problems: 1) The difficulty of find-ing and selectfind-ing appropriate BPs in large collections of BPs impedes the use of BPs, and 2) The low quality of BPDs impedes the use of BPs. Related to these problems, the thesis addresses two sub-goals. The first one is to design a BP Annotation Template for supporting the identification and selection of BPs in BP repositories. The template can be used for organizing and index-ing the contents of BPDs independent of domain. The second sub-goal is to design a BP Document Template for supporting the creation, use and evalua-tion of BP documentaevalua-tion. The BP Document Template offers a structure for describing BPs in a detailed and systematic way.

The research methodology applied is design science, which is the scientific study and creation of artifacts as they are developed and used by people with the goal of solving practical problems. The first artifact, the BP Annotation Template, has been designed and developed based on a literature study and evaluated using expert interviews. The second artifact, the BP Document Template, has been designed and developed based on a literature study using grounded theory as well as on interviews with KM experts. The BP Docu-ment Template has been evaluated using expert interviews and demonstrated by being applied in three real-life cases. By basing the two artifacts on litera-ture studies, the two artifacts consolidate, integrate and extend previous work on BP documentation.

The evaluations indicate that the BP Annotation Template provides a strong foundation for identifying and selecting BPs, independent of domain, and that the BP Document Template can support the structuring of BP docu-ments so that they become complete, uniform and easy-to-use. An identified drawback of the two artifacts is the need for extensive resources for imple-menting and applying them since they may be complex and time-consuming to use. The artifacts are expected to support BP designers as well as BP managers and BP users.

(7)

Acknowledgement

Prima facie, I praise and give thanks to God, the Almighty, for providing me this opportunity, for good health, and for protecting me and granting me the capability to proceed successfully to complete this thesis.

Though only my name appears on the cover of this thesis, it is has been ac-complished in more than five years at the Department of Computer and Sys-tem Sciences (DSV) and a great many people have contributed to its produc-tion and appearance in its current form. I offer my sincere thanks to all of them.

First and foremost, I would like to express my sincere gratitude to my super-visor, Paul Johannesson, for his continuous support of my PhD study and research, for his patience, enthusiasm, motivation, and his immense knowledge. Paul is one of the best teachers that I have had in my life. He sets high standards for his students and he encourages and guides them to meet those standards. Also, I am grateful to my co-supervisor, Erik Perjons, for his encouragement and practical advice. I am thankful to him for facili-tating understanding and communication between Paul and me, commenting on my views, helping me to enrich my ideas. Their excellent and caring guidance helped me through the entire period of the research and the writing of this thesis. We met often, not just during working days, but also during public holidays, summer holidays and after working hours. I cannot imagine having a better supervisor and mentor for my PhD study; one could not wish for a better or friendlier supervisor.

My deepest gratitude goes to my PhD dissertation opponent, Janis Grabis, and the dissertation committee members Mathias Ekstedt, Shengnan Han, William Song and Petros Gougoulakis. Special thanks also go to Shengnan Han, who was my pre-doc opponent and agreed to participate in my pre doc defense committee at the last moment.

I am also indebted to Anneli Edman, Fredrik Bengtsson and Asma Rafiq with whom I interacted during a course of KM in Uppsala University. They introduced me to diverse projects and methods, and participated in my study.

I have been amazingly fortunate to have great people around me from DSV. Particularly, I would like to acknowledge Amin Jalali, Aron Henriksson, Khuram Shahzad, Maria Bergholtz, Maria Skeppstedt, Constantinos

(8)

Gian-noulis, Shahzad Saleem, Iyad Zikra, Hasibur Rahman, Bin Xiao, Eric-Oluf Svee, Parisa Aasi, George Hodosi, Angela Westin, Hercules Dalianis, Rue-ben Laryea, Panagiotis Papapetrou, Jing Zhao, Golrokh Moattar, Nam Aghaee, Robin Eklund, Irvin Homem, Ernest Napoleon, Amer Jwehan, Soren Gustafsson, Thashmee Karunaratne, Amos Zamora, Claudia Ehren-traut, Nasrin Sulemani, Fatima Ferreira, Birgitta Olsson, Love Ekenberg, Lazar Rusu, Jelena Zdravkovic, Sven Olofsson, Lars Asker, Irma Munoz, Josef Swiatycki, Ram Gurung, Martin Henkel, Stewart Kowalski, for the many valuable discussions and gatherings during my journey. Also, I appre-ciate the efforts of the many graduate students at the KM course, who im-plemented my template in real-life cases.

I would like to acknowledge Elias Mturi, Orlando Zacarias, Florence Kivu-nike, Avelino Mondlane, Ranil Colombage, Jean Byungura, Xavier Mui-anga, Constantino Sotomane and Xavier Muianga for numerous discussions during the lunch break. I have also been blessed with friendly and cheerful families and friends who have opened their hearts for me: Alyaa Alfalahi, Ulrika Drougge, Chiara Rossitto, and Khalid Bencherifa. Mohamed El Mekawy has been supportive since the days I began working at DSV. Javier Razuri has helped me get on the road and he has been an experienced mentor at the gym, even though he tortured me.

Many friends have helped me stay sane through these difficult years. Their support and care helped me overcome setbacks and stay focused on my re-search. I greatly value their friendship and I deeply appreciate their belief in me. Special thanks go to Omar Babtaeen, Mohammed Jamalallail, Alex-is MarathovouniotAlex-is, Megbel Alnugaibani, Abdularhim Bukari, Zaid Alshra-ri, Mohammed Algamdi, Ameer Said and Rami Alzahrani.

I also recognize the support, help and wise advice from the PhD Student Ombudsman, Fredrik Ljungqvist, during my case with Harald Kejllin. I am also thankful to Carol de Groot for her excellent editing and proofreading expertise. My sincere thanks also goes to all of the participants in the re-search who have given me insightful comments.

I am so grateful to the Ministry of Health scholarship scheme, which funded my study and the DSV for making it possible for me to study here. My spe-cial and hearty thanks also go to H.E. Hammad Almanea, H.E. Saleh Alshehaib and Abdullah Alderis, who encouraged and supported me.

Last but not least, I would like to thank my sisters, and brothers. They have always supported and encouraged me with their best wishes. Most im-portantly, none of this would have been possible without the love and pa-tience of my mother, Nawal, and my father, Mohsen, my wife, Haifa, my daughter, Aleen and my son, Mansor. They have been a constant source of

(9)

love, concern, support and strength all these years. I would like to express my heart-felt gratitude to my family. They were always there, cheering me up, and they have stood by me through the good times and bad.

I owe my gratitude to all those people who I have mentioned here and others I could not recall. Because of all of these, my study experience has been one that I will cherish forever.

(10)

بﺎﺠﺣ ﻲﻓﺎﺨﻟا ﻦﻣ ﮫﻧود ﺲﯿﻟﺎﻣ ﻲﻠﻟﺎﯾ �ﺎﯾ

رﺪﺘﻘﻣ ﺎﯾ ﺎﻨﺑر ﺎﯾ ﻚﻠﻤﻟا ﻢﯿﻈﻋﺎﯾ

دﺮﻓ ﻚﻧأ ﺪﮭﺷأ

و ءﺎﺟر ﻚﻧأو

أ

بﺎﮭﻣ ﻚﻧ

ﺮﺸﺑ ﺔﺟﺎﺣ ﻲﻓ ﺖﺴﻟ

إ

ﺮﻔﻛ ﻻو ﺮﻜﺷ ن

باﺮﺗ ﺔﻣﻮﻛ ﺮﺸﺒﻟاو ﺰﻋ ﻚﻧأ ﺪﮭﺷأ

بﺎﺑ ﻒﻟأ ﺖﺤﺘﻓ بﺎﺑ ﺮﺸﺒﻟا ﻚﺻﺎﻣ ﻞﻛ

ﺮﺨﺼﻟا تﺮﺠﻓو ءﺎﻤﻟا تﺪﻤﺟ ﻦﻣﺎﯾ ﺖﻧأ

�ﺎﯾ

إ

ﺎﺴﺤﻟا مﻮﯾ ﺎﻨﺑ ﺖﻤﺸﺗﻻ ﻚﻧ

ب

ﺮﻤﻘﻟا ﻖﺸﻨﯾو ﺎﻨﻣ ﺲﻤﺸﻟا ﻲﻧﺪﺗ مﻮﯾ

ﺮﺨﻔﻟا ﻢﻌﻧﺎﯾو ﷲو ﻦﯾﺪﻟﺎﺑ ﺮﺨﻔﻟا

ﻞﻛ ﮫﻟ ﻦﻣﺎﯾ ،ﺪﻤﺤﯾ كﺮﯿﻏ ﺲﯿﻟ اﺪﻤﺣ ﻲﺑرﺎﯾ ،ﺪﻤﺻ ﻚﻧﺄﺑ ﻦﻤﺣرﺎﯾ ﺪﻤﺤﻟا ﻚﻟ

ﺪﺻﻮﯾ ﻻ ﺎﻌﺳاو ﻚﺑﺎﺑ ﺖﯾأرو .. تﺪﺻوأ ﺪﻗ ﻚﻠﻣ ﻞﻛ باﻮﺑأ .. ﺪﻤﺼﺗ ﻖﺋﻼﺨﻟا

.

• " ناﻮﻨﻌﺑ ةﺮﺿﺎﺤﻣ ﻦﻣ ﺲﺒﺘﻘﻣ ﺦﯿﺸﻟا ﺔﻠﯿﻀﻔﻟ "ﮫﻟﻼﺟ ﻞﺟ ﷲ ﺔﻤﻈﻋ ﻲﻧﺮﻘﻟا ﺾﺋﺎﻋ رﻮﺘﻛﺪﻟا - ﷲ ﮫﻈﻔﺣ

(11)

ﻢﯿﺣﺮﻟا ﻦﻤﺣﺮﻟا ﷲ ﻢﺴﺑ

****************

،� ةﺰﻌﻟا ،� ءﺎﯾﺮﺒﻜﻟا ،� ﻚﻠﻤﻟا ،� ﺔﻤﻈﻌﻟا ،� ءﺎﻨﺜﻟا ،� ﺮﻜﺸﻟا ،� ﺪﻤﺤﻟا

� ءﺎﻘﺒﻟا ،� ددﺆﺴﻟا ،� ﺪﺠﻤﻟا

..

ﺪﻤﺤﻟا ﻚﻟو ،ﻦﮭﯿﻓ ﻦﻣو ضرﻷاو تاﻮﻤﺴﻟا رﻮﻧ ﺖﻧأ ﺪﻤﺤﻟا ﻚﻟ ﻢﮭﻠﻟا

أ

ﺖﻧ

مﻮﯿﻗ

ﻦﮭﯿﻓ ﻦﻣو ضرﻷاو تاوﺎﻤﺴﻟا

..

ﺪﻤﺤﻟا ﻚﻟو

أ

ﻦﮭﯿﻓ ﻦﻣو ضرﻷاو تاﻮﻤﺴﻟا ﻚﻠﻣ ﺖﻧ

..

ﺪﻤﺤﻟا ﻚﻟو ﻰﺿﺮﺗ ﻰﺘﺣ ﺪﻤﺤﻟا ﻚﻟ ﻢﮭﻠﻟا

إ

ﺎﺿﺮﻟا ﺪﻌﺑ ﺪﻤﺤﻟا ﻚﻟو ﺖﯿﺿر اذ

..

ا دﻮﻄﻟا ﺐﺣﺎﺻو ،يﺆﻟ ﻲﻨﺑ ﻲﻓ ﺰﻌﻟا ﻞﻣﺎﺣ ﻰﻠﻋ مﻼﺴﻟاو ةﻼﺼﻟاو

ﻲﻨﺑ ﻲﻓ ﻒﯿﻨﻤﻟ

.. فﺎﻨﻣ ﺪﺒﻋ

ﺘﻟاو ةﺮﻐﻟا ﺐﺣﺎﺻ

ﻞﯿﺠﺒ

ﺪﯾﺆﻤﻟا

ﺑ

ﺠ

ﻮﺘﻟا ﻲﻓ رﻮﻛﺬﻤﻟا ،ﻞﯾﺮﺒ

ةار

اﺮﯿﺜﻛ ﺎﻤﯿﻠﺴﺗ ﻢﻠﺳو ﮫﺒﺤﺻو ﮫﻟآ ﻰﻠﻋو ﮫﯿﻠﻋ ﻢﻠﺳو ﷲ ﻰﻠﺻ .. ﻞﯿﺠﻧﻹاو

..

،ﷲ ﻢﻈﻋأو ،ﷲ حﺪﻣأ نأ ﺪﯾرأ ﻚﻟﺬﻟ ،ﺔﺣوﺮطﻹا هﺬھ ﺔﺑﺎﺘﻜﻟ ﻲﻨﻘﻓو يﺬﻟا ﷲ ﺮﻜﺷأ

ﮫﯿﻠﻋ لﺎﻗ ﺎﻤﻛ حﺪﻤﻟا ﻖﺤﺘﺴﯾ ﮫﻧﻷ هﺪﻤﺤﺑ ﻲﺟﺎﻧأ نأ ﺪﯾرأو ،ﷲ ﺢﺒﺳأو ،ﷲ سﺪﻗأو

ﺴﻟاو ةﻼﺼﻟا

ﻊﯾﺮﺳ ﻦﺑ دﻮﺳأ ﻲﺑﺎﺤﺼﻠﻟ مﻼ

"

ﺎﻣأ

نإ

ﻚﺑر

كرﺎﺒﺗ

ﻰﻟﺎﻌﺗو

ﺐﺤﯾ

ﺪﻤﻟا

ح

."

ﺐﺣأ ﺪﺣأ ﻻ مﻼﺴﻟاو ةﻼﺼﻟا ﮫﯿﻠﻋ لﻮﻘﯾو

إ

ﮫﺴﻔﻧ حﺪﻣ ﻚﻟﺬﻟو ﷲ ﻦﻣ حﺪﻤﻟا ﮫﯿﻟ

..

ﺐﯿﺻ ﻞھﺎﻣ ﻦﻤﺣرﺎﯾ ﺪﻤﺤﻟا ﻚﻟ

ﺐﻧﺬﻣ بﻮﺘﻟا ﻞﺒﻘﯾ ﻦﻣﺎﯾ بﺎﺗﺎﻣو

ماﺮﻐﻟا جﺎھﺎﻣ ﺪﻤﺤﻟا ﻚﻟ

مﺎﻤﻐﻟا ﻰﻤھﺎﻣو

بﺮﻄﻤﻟا مﺎﻤﺤﻟا ﻰﻨﻏ ﺎﻣو

ﺪﻤﺤﻟا ،اﺮﯿﺜﻛ اﺪﻤﺣ � ﺪﻤﺤﻟا

ﺪﻤﺤﻟا ،دﺮﯾ ﻻو ﺪﻌﯾ ﻻو ﺪﺤﯾ ﻻ ،ﮫﻟ ﺔﯾﺎﮭﻧ ﻻ اﺪﻤﺣ �

ﻹﺎﺑ �

،ﻦﯿﻘﯿﻟﺎﺑ � ﺪﻤﺤﻟا ،نﺎﻧﺪﻋ ﺪﻟو ﺪﯿﺴﺑ � ﺪﻤﺤﻟا ،نآﺮﻘﻟﺎﺑ � ﺪﻤﺤﻟا ،نﺎﻤﯾ

� ﺪﻤﺤﻟا

ﺔﻤﻌﻨﺑ

ﻦﻣﻷا

ﺔﯿﻓﺎﻌﻟاو

،

ﻞﻛ ﻰﻠﻋ � ﺪﻤﺤﻟا

ﮫﻤﻌﻧ

..

ﺐﺋﺎﺧ ﻦﻣﺆﻤﻟﺎﻓ ﻻإو ﻚﻨﻣو بﺈﻛﺮﻟا ﺪﺸﺗﻻ ﻻإو ﻚﯿﻟإ

ﻀﻣ ماﺮﻐﻟﺎﻓ ﻻإو ﻚﯿﻓو

بذﺎﻛ ثﺪﺤﻤﻟﺎﻓ ﻻإو ﻚﻨﻋو ﻊﯿ

ﻰﻟﺎﻌﺗو ﮫﻧﺎﺤﺒﺳ ﮫﻟ ﺔﯾﺎﮭﻧﻻ اﺮﯿﺜﻛ اﺪﻤﺣ ﮫﻤﻌﻧ ﻰﻠﻋ ﷲ ﺪﻤﺣأ

..

ﻹا ﻢﯾﺪﻗ ﷲو نﺎﺴﻠﺑ ﻞﻜﺑ ﺪﻤﺤﯾ ﷲ

ﻞﻜﺑ ﺪﻤﺤﯾ ،نﺎﺴﺣ

ﺔﺠﮭﻟ

ﻐﻟ ﻞﻜﺑ ﺪﻤﺤﯾ ،

.. ﺔ

(12)

Abbreviations

AISeL Association for Information Systems Electronic Library

BP Best Practice

BPD Best Practice Document

BPMS Business Process Management System

BSC Balanced Scorecard

ERP Enterprise Resource Planning

IS Information System

IT Information Technology

KM Knowledge Management

KS Knowledge Sharing

(15)

List of Figures

Figure 1.The relations between the research papers and problems and goals of the thesis ... 23 Figure 2. Dimensions of KM adapted from (Jashapara, 2011) ... 29 Figure 3. Spiral Model of the Knowledge Conversion and

Self-transcending Process (Nonaka and Konno, 1998) ... 30 Figure 4. Processes for a KM, adapted from Dalkir (2011) ... 32 Figure 5. Four steps of BPs in an organization database, adapted

from (American Productivity and Quality Centre, 1997) ... 35 Figure 6. Activities and their outputs of the design science process,

adopted from Johannesson and Perjons (2014) ... 44 Figure 7. Research process for the BP Annotation Template... 45 Figure 8. The activities in the development of the BP Annotation

Template ... 47 Figure 9. Research process for the design of the BP Document

Template ... 50 Figure 10. The activities to design and develop the BP Document

Template ... 52 Figure 11. Five phases and included steps in grounded theory

method for reviewing the literature, adopted from Wolfswinkel et al. (2013) ... 53 Figure 12. A conceptual model of BP concepts. The concepts

needed for an unstructured description of BPs are marked with a gray background ... 58 Figure 13. A conceptual model of BP concepts. The concepts

needed for a structured description of BPs are marked with a gray background ... 59 Figure 14. A conceptual model of BP concepts including metadata

(16)

Figure 15. The research strategy and methods applied in the evaluation of the BP Annotation Template ... 79 Figure 16. Aggregate of the answers across all of the BP values

regarding the easy-to-annotate requirement ... 82 Figure 17. Aggregate of the answers across all BP attributes

regarding the easy-to-annotate requirement ... 83 Figure 18. Aggregate of the answers across all of the BP values

regarding the ‘applicable to any BP’ requirement ... 84 Figure 19. Aggregate of the answers across all BP attributes

regarding the ‘applicable to any BP requirement’ ... 85 Figure 20. Aggregate of the answers across all of the BP values

regarding the ‘domain-independent requirement’ ... 86 Figure 21. Aggregate of the answers across all BP attributes

regarding the ‘domain-independent’ requirement ... 87 Figure 22. The research strategy and methods applied in the

evaluation of the BP Document Template ... 91 Figure 23. The overall assessed average value for 33 attributes in

(17)

List of Tables

Table 1. The relationships between data, information and

knowledge ... 28

Table 2.Different definitions of KM by different researchers ... 29

Table 3. Different processes for KM in the literature ... 31

Table 4. Different definitions of the term, Knowledge Sharing ... 33

Table 5. A fictitious example of BP ... 37

Table 6. Inclusion and exclusion criteria ... 48

Table 7. Inclusion and exclusion criteria ... 49

Table 8. Inclusion and exclusion criteria for article search ... 53

Table 9. Inclusion and exclusion criteria for article selection ... 54

Table 10. Showing the number of time a value exist in literature .. 66

Table 11. The BP Annotation Template ... 68

Table 12. Showing the number of times an attribute occurs in literature ... 73

Table 13. The final BP Document Template ... 74

Table 14. A fictitious example of applying the BP Document Template ... 77

Table 15. Benefits of the BP Annotation Template ... 89

Table 16. Drawbacks of the BP Annotation Template ... 89

Table 17. Suggested changes to the BP Annotation Template ... 90

Table 18. Assessment result for each BP attribute of the qualitative analysis ... 94

Table 19. Benefits of BP Document Template ... 98

Table 20. Drawbacks of BP Document Template ... 99

(18)

(19)

19

1. Introduction

This chapter describes the background to the thesis, the problems addressed, the research question, and the goals of the research.

1.1. Background

As institutions and societies become ever more complex, organizations need to develop strat-egies for identifying, creating, sharing and applying their knowledge assets by means of Knowledge Management (KM) (Dalkir, 2011). Snowden (2002) have identified three genera-tions of KM. The first generation focused on containers of knowledge demonstrated by the phrase “if only we knew what we know”, and by introducing IT approaches in KM. Organiza-tions then started to utilize their knowledge assets more effectively by implementing internal KM systems and intranets. This second generation emphasized the importance of people and cultural issues in KM. The focus here was on people, and organizations realized the im-portance of the bottom-up adoption of KM by using IT solutions. The third generation is about the awareness of shared context, which creates shared meaning. It is about describing and organizing content so that end users can easily use and apply it. This phase is character-ized by content management and the advent of metadata to describe the content. We are now in this third generation according to Snowden (2002). Knowledge sharing is the focus of this thesis.

Knowledge sharing means the provision of know-how and other types of knowledge to help employees to cooperate with others to develop new ideas, solve problems or apply procedures or policies to make the organization more effective and efficient (Wang and Noe, 2010). Knowledge sharing is critical to an organization’s success as it leads to the acceleration of the deployment of knowledge to parts of the organization that can significantly benefit from it (Syed-Ikhsan and Rowland, 2004). Therefore, facilitating access to better knowledge throughout the organization helps employees to adopt and innovate their practices and deliver better quality work and products (Wiig, 2004). Knowledge sharing may happen via face-to-face interactions, through apprenticeship, via written correspondence or documents, by carry-ing out organizational routines and processes or by applycarry-ing technologies in which knowledge is embedded (Wang and Noe, 2010; Nonaka and Takeuchi 1995; Argote 2013). One of the most widely used means to share knowledge is via Best Practice (BP). A BP can be defined as “the most efficient (least amount of effort) and effective (best results) way of

accomplishing a task, based on repeatable procedures that have proven themselves over time for large numbers of people” (cited from Wikipedia in Graupner et al. (2009)). For the past

two decades, the use of BP to share knowledge has been a popular means to move organiza-tions towards higher performance in order to be more successful (Whittle et al. 1992; Szulan-ski, 1996; O’Dell and Grayson, 1998; Davies and Kochhar, 2002; Netland and Alfnes, 2011; Watson, 2007). As an example, Cormican and O’Sullivan (2004) proposed a BP approach to identify the most valuable and successful practices in order to develop an effective product development process, which would result in competitive advantages for the organizations introducing the BPs.

Sharing BPs can affect a company’s performance in a number of ways, such as an increased return on investment, increased value added per employee, and improved customer

(20)

satisfac-20

tion (Goodman and Goldman, 2007; Pfeffer and Sutton, 2006; Gold et al. 2001). Reddy and McCarthy (2006) stated some benefits that organizations may gain from the effective sharing of their BP documents: “1) identifying and replacing of poor practices; 2) raising the perfor-mance of poor performers to closer to that of the best; 3) avoiding ‘reinventing the wheel’; 4) minimizing duplication of work caused by the use of poor methods; 5) saving costs through better productivity and efficiency”.

However, there is a risk that organizations’ BP initiatives can fail (Barrett and Stanley, 1999; Whittle et al.1992, Davies and Kochhar, 2000). Two main challenges in BP application can be summarized as follows:

• the difficulty of finding and selecting appropriate BPs in large collections of BPs (Simard and Rice, 2007; Dani et al. 2006; Mansar and Reijers, 2007; Hanafizadeh et al. 2009);

•_{the low quality of Best Practice Documents (BPDs), for example, a lack of} infor-mation of how the BPs actually work in organizations as well as their usefulness (Da-na and Smyrnios, 2010).

1.2. Problem Definition

This research focuses on organizing BPDs, in line with the above-mentioned characteristics of the third generation of KM. Therefore, the overall problem that this research aims to ad-dress is that it is challenging to share and use BPDs in organizations.

This thesis focuses on two practical problems that are root causes of this overall problem.

Practical problem 1

The difficulty of finding and selecting appropriate BPs in large collections of BPs im-pedes the use of BPs.

It is a well-known problem that it is difficult to find and select BPs in large collections or repositories of BPs (Simard and Rice, 2007; Dani et al. 2006; Mansar and Reijers, 2007; Hanafizadeh et al. 2009). Practitioners do not know how to find appropriate BPs and which BPs to select among several similar ones, as argued by Abd Rahman et al. (2011). Instead of searching in the repository for BPs, practitioners tend to revert to informal communication channels and centralized authority to find ways of working or solutions to certain problems. These practices may not be the most efficient and effective ones.

Two main factors make it difficult to find and select appropriate BPs. There is a lack of a common terminology used by both practitioners and in BP documents, and there is a lack of domain-independent search indices.

The first factor is caused by the fact that practitioners use another terminology in their daily work than the one found in existing BPs, including their metadata, That is, the terminology used in BPs as well as their metadata, does not correspond to the one that is used by practi-tioners in their work environment (Dourish et al. 2000; Tough and Moss 2003). Hence, it is difficult for them to use the metadata, or navigate and understand the content of the BPs in order to find the appropriate ones (Kao et al. 2003; Mas and Marleau, 2009).

The second factor is the lack of domain-independent search indices (Vesely, 2011; Smith et al. 2010; Zhu et al. 2007; Graupner et al. 2009), i.e., search indices that consist of terms that are not associated with a specific domain. The lack of such indices makes it difficult for prac-titioners to find BPs in other domains then the one with which they are familiar. This is a drawback because practitioners may need to find BPs from other domains than those with

(21)

21

which they are familiar. For example, in small and medium size organizations, practitioners may need to manage several domains, and they may not be familiar with all of them. Another example is a high level manager that may want to obtain an overview of what BPs exist re-garding a certain aspect of his organization that is independent of a particular domain. A third example is when there is a need to integrate BPs from different domains in order to carry out a certain task (Simard and Rice, 2007).

This practical problem is also related to a knowledge gap in the research literature, which has further motivated the research carried out and presented in this thesis.

Knowledge gap 1

There is a lack of knowledge about IS artifacts for finding and selecting appropriate BPs in BP repositories.

Vesely (2011) has emphasized the lack of instruments for selecting and finding BPs. Vesely states that practitioners tend to select BPs randomly, subjectively and without proper justifica-tion. Smith et al. (2010) point out that the selecting and finding BPs is affected by so-called confirmation bias; that is, the practitioners search and select BPs that confirm their own be-liefs.

Practical problem 2

The low quality of BPDs impedes the use of BPs.

The practical problem stated above uses the expression “low quality of BPDs”. In this thesis, this means that the BPD is not fit for its purpose. Some underlying causes for the low quality of documentation, in general, are that it has missing, incomplete, non-clear, redundant, incor-rect, inconsistent, or irrelevant content.

Low quality BPDs means, for example, that the purpose and value of a BP is not clearly stat-ed; that the description of the BP is not detailed enough to be easily applistat-ed; and/or that the earlier experiences of applying the BP are not presented.

Low quality BPDs lead to situations in which practitioners are not able to correctly and effi-ciently use BPs, or may not trust them. Hence, low quality BPDs can prevent the successful use BPs as a means of knowledge sharing. Researchers have emphasized that the lack of un-derstanding of the purpose of a BP and the failure to measure the value of the knowledge within a BP are major barriers to the successful management of knowledge (Tabrizi et al. 2011; Aggestam and Persson, 2010; Dyer and McDonough, 2001).

The research presented in this thesis focuses on one aspect of low quality, that is, incomplete-ness. More precisely, the focus is on the lack of context in documentation. Existing BPDs are often static and do not provide appropriate context to enable the successful application of BP (Kothari et al. 2011). More precisely, a BP is bound to a specific set of circumstances and conditions to achieve its expected result. So far, the context of BP applications has only at-tracted limited theoretical attention. Dinur et al. (2009) argued that “When context is

dis-cussed, it is mostly concerned with the context of the person possessing the knowledge or the human context surrounding the knowledge itself. In addition, the contextual embeddedness of knowledge is usually discussed on an organizational level of analysis” (Dinur et al. 2009). It

is this “contextual embeddedness of knowledge on an organizational level” supporting BP applications, that is a focus of our research.

(22)

22

Another important focus in our research is a lack of clarity, that is, a lack of precise, concise and easily understandable documentation. According to Hall (2002) and Rainey (2003), clear documentation helps to describe as well as maintain organizational practices.

The practical problem is also related to a knowledge gap in the research literature, which has further motivated the research carried out and presented in this thesis.

Knowledge gap 2

There is a lack of knowledge about IS artifacts for supporting the creation and evalu-ation of BPDs.

Kao et al. (2003) and Dani et al. (2006) have emphasized the need for having instruments to enhance the quality of BPD, thereby enabling better search quality and more use of BPDs. Without properly documented BPs, it is difficult to share them within an organization.

1.3. Research Question and Goals

The overall research question in this thesis is:

How should IS artifacts be designed in order to improve the use of BPs in organiza-tions?

Within the IS area, there are various artifacts that range from software architectures, formal system, enterprise architecture, information models, and design guidelines to demonstrators, and production systems. An artifact is an object that has been made by humans with the pur-pose of addressing a practical problem. Stakeholders want to use the artifact in order to solve the practical problem or at least alleviate it. Thus, IS artifacts are always embedded in the context of a larger problem and do not exist in isolation, (Johannesson and Perjons, 2014). The overall goal of the research presented in this thesis is:

to design IS artifacts for improving the use of BPs in organizations.

This goal is achieved by addressing two of its sub-goals, which correspond to the two practi-cal problems defined in section 1.2.

The first sub-goal is:

to design a BP Annotation Template for supporting the identification and selection of BPs in BP repositories.

The BP Annotation Template is a structure for describing BPs in a concise and high level way. The template can be used for organizing and indexing the contents of BPDs in a do-main-independent way. To use the BP Annotation Template is one solution to handle this problem. However, there are also other solutions for addressing the problem, including ontol-ogy, semantic search and intelligent system (Fensel, 2003; Baziz, 2004). These approaches have not been directly addressed in this thesis, but we believe that the annotation template solution can be fruitfully combined with them. In particular, the BP Annotation Template can subsequently be used to improve semantic search results by highlighting essential attributes of BP documents that can help to navigate and retrieve BP documents. In particular, the thesis focused solely on annotating attributes of BP documents, which can be reliably extracted from BP documents, and in exploring alternative ways in which this semantic information may be exploited by the BP Annotation Template to improve retrieval performance.

(23)

23 The second sub-goal is:

to design a BP Document Template for supporting the creation, use and evaluation of BPDs.

The BP Document Template is a structure for describing BPs in a detailed and systematic way. The template can help knowledge engineers to develop high quality BPDs before dis-seminating them. The template can also be used to assess existing BPDs in order to enhance them.

1.4. Structure of the Thesis and Publications

The thesis follows the Gregor and Hevner (2013) schema for design science research. Hence, the thesis starts with an introduction in the first chapter, followed by a review of related work. The third chapter presents the research methodology, while the fourth chapter includes de-scriptions of the artifacts. Evaluations are described in Chapter 5 and the conclusions and discussion are given in Chapter 6. The main contributions of this thesis have been published in five included papers, as shown in Figure 1, and described below.

Figure 1.The relations between the research papers and problems and goals of the thesis

• Paper 1. Alwazae, M. Kjellin, H. and Perjons, E. (2014). A synthesized classification system for BPs. VINE: The Journal of Information and Knowledge Management Sys-tems, 44 (2), 249-266.

Meshari Alwazae’s contribution to the paper corresponds to more than 60 percent. He was the first author of the paper and he contributed to all parts, including the identification of the arti-cles to be included in the reference artiarti-cles in order to identify BP attributes and the design of

(24)

24

a synthesized classification system for BPs. Harald Kjellin and Erik Perjons provided contin-uous feedback on all the contributions, which significantly improved the paper. Also, they reviewed the paper and improved the writing.

• Paper 2. Alwazae, M.M.S. Johannesson, P. and Perjons, E. (2015). Evaluation of a classification system for best practices. 48th Hawaii International Conference on Sys-tem Sciences (HICSS), 3702-3711.

Meshari Alwazae’s contribution to the paper corresponds to more than 70 percent. He was the first author of the paper and he contributed to all parts including the development of the clas-sification system for BPs. Paul Johannesson contributed to the clasclas-sification system with two BP attributes with their allowed values. Paul Johannesson and Erik Perjons reviewed the pa-per and improved the writing.

•_{Paper 3. Alwazae, M.M.S. Perjons, E. and Kjellin, H. (2014). Quality measures for} documentation of best practices. 47th Hawaii International Conference on System Sciences (HICSS), 3410-3419.

Meshari Alwazae’s contribution to the paper corresponds to more than 70 percent. He was the first author of the paper and he contributed to all parts including the development of quality measures for documentation of BPs. Erik Perjons was involved in several discussions and continuous feedback on the quality measures for documentation of BPs and their use for im-proving BP documents. Also he reviewed the paper several times and significantly improved the writing. Harald Kjellin conducted one interview with a respondent.

•_{Paper 4. Alwazae, M. Perjons, E. and Johannesson, P. (2015). Template-driven best} practice documentation. Submitted.

Meshari Alwazae’s contribution to the paper corresponds to more than 70 percent. He was the first author of the paper and he contributed to all parts, including identification of the articles to be included in the reference articles in order to identify BP attributes and designing the BP Document Template. He also carried out the evaluation and data analysis. Paul Johannesson and Erik Perjons were involved in several discussions and continuous feedback on the design process for the template and its use for improving BP documents. Also, they reviewed the paper and significantly improved the writing.

•_{Paper 5. Alwazae, M. Perjons, E. and Johannesson, P. (2015). Applying a template} for best practice documentation. The 3rd_{Information Systems International}

Confer-ence, November, 2-4, Surabaya, Indonesia.

Meshari Alwazae’s contribution to the paper corresponds to more than 70 percent. He was the first author of the paper and he contributed to all parts, including identification of the case study, and the respondents, and he also conducted the interviews. Erik Perjons and Paul Jo-hannesson provided continuous feedback on the paper. Also, they reviewed the paper and significantly improved the writing.

Below is a list of publications the author has contributed to, but which are not included in the thesis.

•_{Alwazae, M. Perjons, E. and Kjellin, H. (2013). Verifying the Usefulness of a} Classi-fication System of Best Practices. The 5th_{International Conference on Knowledge}

Management and Information Sharing, Vilamoura, Algarve, Portugal, SciTePress, 405-412.

(25)

25

•_{Alwazae, M. and Kjellin, H. (2013). Creating Best Practices in Saudi Arabia. In} Ad-vances in Business Related Scientific Research Journal (ABSRJ), Edukator, 4 (2), 113-125.

•_{Alwazae, M. and Kjellin, H. (2012). An empirical Investigation of how Best Practices} can be Shared, Knowledge Management Middle East, March 12-13, Academic Con-ferences Limited.

• Alwazae, M. and Kjellin, H. (2012). An empirical investigation of how best practices can be created a case study from Saudi Arabia. In Advances in Business Related Sci-entific Research Conference (ABSRC), September 5-7, Olbia, Italy.

• Kjellin, H. and Alwazae, M. (2011). Automated feedback to facilitate the understand-ing of filmed best practices. In Proceedunderstand-ings of 4th_{Conference of e-Learning}

Excel-lence in the Middle East, January 31-February 3, Dubai, United Arab Emirates, Inter-national council for open and distance education. 339-348.

(26)

(27)

27

2. Related Work in KM and BPs

This chapter describes the research setting for the artifacts developed, that is, the BP Annota-tion Template and the BP Document Template. First, the KM background is described, in-cluding the notion of knowledge sharing. Second, BP is defined along with examples of vari-ous BP systems and solutions as well as identified challenges for BPs. Third, key concepts and existing solutions for annotating BPs are discussed. Finally, key concepts and existing solutions for documenting BPs are discussed.

2.1. KM Background

This section introduces the KM background that underlies this thesis. The notions of data, information and knowledge are introduced in the first part. Second, KM is defined, and third, KM processes are presented. Finally, knowledge sharing is discussed.

2.1.1. Classification of Knowledge

To clarify what knowledge for organizations might mean, it is helpful to differentiate between data, information and knowledge. The relations between data, information and knowledge (sometimes also including understanding and wisdom) can be defined in different ways, as shown in Table 1. As the table shows, knowledge is usually defined as some sort of combina-tion between informacombina-tion and personal experience, and therefore hard to transfer between people.

Researchers Definitions

Ackoff (1989) •_• Data: Symbols _{Information: Data that are processed to be useful; provides} answers to who, what, where, and when questions

• Knowledge: Application of data and information; answers how questions

• Understanding: Appreciation of why • Wisdom: Evaluated understanding Choo et al. (2000) •_• Data: Facts and messages _{Information: Data vested with meaning}

•_{Knowledge: Justified, true beliefs}

Liew (2007) • “Data are recorded (captured and stored) symbols and _{signal readings.”} •_{“Information is a message that contains relevant meaning,}

implication, or input for decision and/or action. Infor-mation comes from both current (communication) and his-torical (processed data or ‘reconstructed picture’) sources. In essence, the purpose of information is to aid in making decisions and/or solving problems or realizing an oppor-tunity.”

•_{“Knowledge is the (1) cognition or recognition} (know-what), (2) capacity to act (know-how), and (3) understand-ing (know-why) that resides or is contained within the mind or in the brain. The purpose of knowledge is to better

(28)

28

our lives. In the context of business, the purpose of knowledge is to create or increase value for the enterprise and all its stakeholders. In short, the ultimate purpose of knowledge is for value creation.”

Table 1. The relationships between data, information and knowledge

The description of data, information, and knowledge as a hierarchy of higher and higher val-ue, as shown in the examples in Table 1, has been criticized by, for example, Alavi and Leidner (2001). The basis for the transformation from data to information to knowledge is often a mixture of different dimensions at the same time, such as context, usability and inter-pretability. This makes it hard to determine when, for example, information is converted to knowledge in a certain situation. Instead, Alavi and Leidner (2001) suggest the following distinction: “Information is converted to knowledge once it is processed in the mind of indi-viduals and knowledge becomes information once it is articulated and presented in form of text, graphics, word and other symbolic forms” (Alavi and Leidner, 2001). This thesis adopts the distinction made by Alavi and Leidner (2001). It is also based on the assumption that knowledge can be shared, at least partly, via information but also via observation, imitation and guidance.

The distinction between information and knowledge is also, according to Alavi and Leidner (2001), closely related to the distinction between explicit knowledge and tacit knowledge. Explicit knowledge is expressed in the form of text, numbers, codes, and formulae, while tacit knowledge is situated in the minds of people and, therefore, it is often difficult to formulate in an explicit way. While it is relatively easy to transfer explicit knowledge through formal lan-guage, it is much more difficult to transfer tacit knowledge (Nonaka and Takeuchi 1995). While the field of information management is about managing explicit knowledge or infor-mation in organizations, KM is about managing and transferring both tacit and explicit knowledge (Frappaolo, 2006), and managing the conversion activities between the two types of knowledge (Nonaka and Takeuchi 1995).

2.1.2. Definition of KM

KM is an interdisciplinary field and has its roots in different disciplines, as shown in Figure 2. The most common disciplines for publishing KM literature are within IS and Human Re-source Management (Jashapara, 2011). However, this interdisciplinary status gives KM valu-able possibilities for synergy when carrying out a dialogue with other disciplines. For exam-ple, Guo and Sheffield (2008) have proposed the following fields for a paradigmatic and methodological examination of KM research: IS, Management and Systems Thinking.

(29)

29

Figure 2. Dimensions of KM adapted from (Jashapara, 2011)

Given the multidisciplinary fields of KM, KM is defined differently by different authors in the literature. Table 2 presents a number of alternative definitions of KM. The definitions offered by Swan et al. (1999), and Mertins et al. (2003) focus on the activities of creating, sharing and applying knowledge; Mertins et al. (2003) from a IS perspective; Swan et al. (1999) from a human resource perspective. Beijerse (2000) and Newell et al. (2009) relate KM to strategies and policies used in organizations, while North and Kumta (2014) empha-size both strategic and operational objectives.

Researchers Definitions

Swan et al. (1999) “.. any process or practice of creating, acquiring, capturing, sharing,

and using knowledge, wherever it resides, to enhance learning and performance in organisations”

Mertins et al.

(2003) “… all methods, instruments and tools that in a holistic approach contribute to the promotion of core knowledge processes”

Newell et al.

(2009) “… improving the ways in which firms facing highly turbulent envi-ronments can mobilise their knowledge base (or leverage their knowledge ‘assets’) in order to ensure continuous innovation”

Beijerse (2000) “The achievement of the organisation’s goals by making the factor

knowledge productive”

North and Kumta

(2014) “Knowledge management enables individuals, teams and entire or-ganisations as well as networks, regions and nations to collectively and systematically create, share and apply knowledge to achieve their strategic and operational objectives. Knowledge management contributes to increase the efficiency and effectiveness of operations on the one hand and to change the quality of competition (innova-tion) on the other by developing a learning organisation”

Table 2.Different definitions of KM by different researchers

Nowadays, KM is oriented to organizational objectives that include integration, innovation and the creation of competitive advantage, improved performance, and the sharing of BP (Bennis and O’Toole, 2005; Ghoshal, 2005; Serenko et al. 2008). Therefore, this thesis adopts Jashapara’s definition of KM: “the effective learning process associated with exploration,

(30)

tech-30

nology and cultural environments to enhance an organization’s intellectual capital and per-formance” (Jashapara, 2011). Jashapara’s definition is based on interdisciplinary perspectives,

and it gathers and brings different dimensions of KM together.

2.1.3. Knowledge Management Processes

Within the last two centuries, organizations have realized the important assets of both tacit and explicit knowledge for their success (Davenport and Prusak, 1998; Alavi and Leidner, 2001; Lee et al. 2014). One of the most influential efforts in KM is the work that described knowledge creation, conversion and transformation as a knowledge spiral model. See, for example, (Nonaka and Takeuchi 1995) and (Nonaka and Konno, 1998). The model empha-sizes that organizational knowledge develops through dynamic and continuous interaction between explicit and tacit knowledge. The model describes four steps in the processes of knowledge transformation or conversion, as shown in Figure 3.

Figure 3. Spiral Model of the Knowledge Conversion and Self-transcending Process (Nonaka and Konno, 1998)

1. Socialization

Socialization is the process of capturing tacit knowledge from another person through direct interaction via dialogue, observation, imitation or guidance.

2. Externalization

Externalization is the process that translates or converts tacit knowledge into explicit knowledge. Externalization is supported by the articulation of tacit knowledge and the trans-lation of it into an understandable form.

3. Combination

Combination is the process of putting together different bodies of explicit knowledge into more complex sets of explicit knowledge, e.g., when people exchange knowledge and com-bine explicit knowledge. This phase relies on three processes; first, collecting and integrating new explicit knowledge; second, disseminating explicit knowledge by means of meetings or presentations; third, processing and editing explicit knowledge to make it more usable.

(31)

31 4. Internalization

Internalization is the process of transforming explicit knowledge into tacit knowledge. In practice, two factors support this phase. First, explicit knowledge has to be demonstrated in practice and action. Second, the process of demonstrating the explicit knowledge has to en-gage ‘learning by doing’.

As mentioned above, the spiral model emphasizes that organizational knowledge develops through dynamic and continuous interaction between explicit and tacit knowledge. Other researchers emphasize other activities or processes for KM. Table 3 summarizes.

Author Processes/ Activities

Meyer and Zack (1996) Acquire, refine, store, distribute and present knowledge Alavi and Leidner (2001) Create, store and /or retrieve, transfer and apply

knowledge

Tiwana (2000) Analyze, align, design, audit, design, create, develop, deploy, manage and evaluate knowledge

Awad and Ghaziri (2004) Capture, organize, refine and transfer knowledge Sagsan (2009) Create, share, structure, use and audit knowledge

Turban et al. (2011) Create, capture, refine, store, manage and disseminate knowledge

O’Dell et al. (2003) Organize, share, adapt, use, create, define and collecting knowledge

Bukowitz and Williams

(2000) Get, use, learn, contribute, assess, build and/or sustain and divest knowledge Wiig (1994) Build, hold, pool and apply knowledge

Tiwana (2000) Analyze, align, design, audit, design, create, develop, deploy, manage and evaluate knowledge

Table 3. Different processes for KM in the literature

The above processes are neither exhaustive nor mutually exclusive. Dalkir (2011) proposed a set of common processes for KM. These processes are knowledge capture and/or creation, knowledge sharing and dissemination, and knowledge acquisition and application, as shown in Figure 4. The definitions of the processes are as follows:

•_{Knowledge capture and creation - Knowledge capture is to identify existing internal} knowledge in an organization, and knowledge creation is to develop new knowledge; • Knowledge sharing and dissemination - Knowledge sharing is the movement of

knowledge throughout the organization to get greater innovation and reuse for greater efficiency, and knowledge dissemination is the deliberate leveraging of knowledge as well as enabling knowledge to be spread via organizational learning and memory; • Knowledge acquisition and application - Knowledge acquisition is “the process of

ex-tracting, transforming and transferring expertise from a knowledge source” (Dalkir,

(32)

32

Figure 4. Processes for a KM, adapted from Dalkir (2011)

This thesis focuses on the knowledge-sharing process/activity, although other processes pro-posed by Dalkir (2011) are also of importance. Sharing knowledge is one of the most funda-mental challenges in KM and important for its success (Jashapara, 2011).

2.1.4. Knowledge Sharing

Knowledge Sharing (KS) between employees and within and across teams is fundamental for successful KM and innovation (Jackson et al., 2006). According to Wang and Noe (2010), research has “shown that knowledge sharing and combination is positively related to reduc-tions in production costs, faster completion of new product development projects, team per-formance, firm innovation capabilities, and firm performance including sales growth and rev-enue from new products and services”.

Many different definitions of KS exist, as shown in Table 4. Some of them emphasize the need for the recipient to re-construct the knowledge shared. Related to the term KS are also the terms, knowledge transfer and knowledge exchange. These three terms are sometimes used as synonyms, but sometimes not. For example, the term “knowledge transfer” is some-times used to emphasize the movement of knowledge between different sections, departments or organizations rather than between individuals (Ipe, 2003; Szulanski et al. 2004). Moreover, the term “knowledge exchange” sometimes includes actors providing knowledge to each oth-er (as in KS) as well as employees seeking knowledge from each othoth-er (i.e., knowledge seek-ing) (Wang and Noe, 2010; Cabrera et al. 2006).

Author Processes/Activities

Wang and Noe (2010) “Knowledge sharing refers to the provision of task information and know-how to help others and to collaborate with others to solve problems, develop new ideas, or implement policies or pro-cedures”.

Maier et al. (2009) KS is the process of one person (source) (1) deciding to share knowledge, (2) remembering a portion of knowledge, (3) expli-cating it to contextualized information on a medium, (4) actively or passively transferring it to another person (recipient) who (5) perceives the information and (6) interprets it in the given context so that the knowledge is re-constructed and integrated in the per-son’s knowledge base. (7) Finally, newly acquired knowledge is evaluated by the recipient.

Wikipedia (2015) “Knowledge sharing is an activity through which knowledge (namely, information, skills, or expertise) is exchanged among

(33)

33

people, friends, families, or communities (for example, Wikipe-dia), or organizations.”

Ipe (2003) “Knowledge sharing is basically the act of making knowledge available to others within the organization. Knowledge sharing between individuals is the process by which knowledge held by an individual is converted into a form that can be understood, absorbed, and used by other individuals.”

Table 4. Different definitions of the term, Knowledge Sharing

Renzl and his colleagues (2006) proposed two different approaches to the sharing of knowledge: (1) approaches mainly for sharing tacit knowledge such as communities of prac-tice and storytelling; and (2) approaches mainly for sharing explicit knowledge, such as Pat-terns and BP databases.

Community of Practice

A community of practice refers to a work-related group of people who share common prob-lems or interests, and who meet informally to learn from each other through ongoing interac-tions (Wenger and Snyder, 2000; Prusak and Matson, 2006).

A community of practice includes the advantage that it can solve problems, promote the shar-ing of BPs, develop the professional skills of employees, create a new line of business, and help the organization retain and recruit talent (Prusak and Matson, 2006).

A main drawback with a community of practice is that it is not easy to build and sustain with-in an organization because of its with-informal nature. The participants can be unwillwith-ing to accept interference and supervision, and are therefore hard to manage (Prusak and Matson, 2006).

Storytelling

Storytelling is one of the approaches to share knowledge that was introduced to organizations a long time ago. It has been used by organizations to share norms and values, develop trust and commitment in the organization, share tacit knowledge, facilitate unlearning and generate emotional connection. Thus, storytelling is an effective tool for sharing knowledge within communities of practice (Sole and Wilson, 2002; Dalkir, 2011).

Storytelling plays significant roles in capturing, storing and utilizing tacit knowledge in or-ganizations, but can also play a role in storing and using explicit knowledge if the story is written down. Recently, the use of storytelling has grown rapidly among organizations as a deliberate tool to share knowledge, and it is the preferred method among some management consultants. The reason for this is that is uses authentic language to inspire, engage, attract and involve people in a narrative, in a fun and interesting form (Denning, 2001).

Based on the Knowledge Management Toolkit (2009), the benefits of storytelling are that it can:

• communicate and illustrate how to apply practices to solve challenges but also specif-ics of a context, that is, it can describe context specific knowledge;

• create changes in behavior and attitudes;

•_{create shared understanding among people about future direction and ambition.} The drawbacks of storytelling are that stories represent one perspective, i.e., a single point-of-view, which may be less relevant to other individuals. Moreover, the impact of storytelling depends on whether it is shared in a written or an oral form, and who is the author or teller. A limitation is also the fact that it does not replace analytical thinking, which some might think (Sole and Wilson, 2002).

(34)

34

Pattern

A pattern describes a solution to a specific problem, which recurs in a variety of cases in the context of an organization. Hence, each pattern combines a problem with a solution, reflect-ing the context and explainreflect-ing how the pattern can be applied. A pattern has four general el-ements: pattern name, problem, solution, and the consequences of applying the pattern (Gamma et al. 1994). Riehle and Zullighoven (1995) define pattern as “an abstraction from a

concrete form, which keeps recurring in specific non-arbitrary contexts”. In order to be

ap-plied in a variety of cases, a pattern has an abstract form and it can then be adapted to fit the situation at hand.

The term ‘pattern’ has been widely used in software development, building architectural de-sign (often inspired by the pattern language of Alexander et al. (1977)) and workflow man-agement (Stephenson and Bandara, 2007). It has also been discussed in the area of organiza-tional culture regarding teaching new employees the correct way to think, perceive and feel in relation to problems within the organizational context (Maier, 2007).

A pattern can be developed based on theoretical or practical evidence. Thus, a pattern is not invented. It is rather discovered by theoretical evidence or observation.

The differences between a pattern and a BP can be summarized as follows:

• a pattern can be built solely based on theoretical evidence, while a BP cannot;

• a pattern is often documented in a formal template, while a BP is often documented informally, for example, by using free text formats with little structure (Graupner et al. 2009);

• a pattern has a much smaller focus and can be independent, and can be seen as a building block that needs to be combined with other patterns in order to form a mean-ingful entity, while each BP is an independent entity (Stephenson and Bandara, 2007).

BP Database

A BP database is a database of BPs and needs to provide enough information for a potential user of the BP to find it and to be able to decide if it is worth applying for the situation at hand (Renzl et al. 2006).

2.2. Best Practice

This section starts with definition and discussion of the notion of BP followed by presenting cases of BP implementations. Challenges for applying BPs is presented at the end of this sec-tion.

2.2.1. Definition and Discussion of the Notion of BP

Even though the merits of BP are generally agreed upon, the scope, context and definition of BP are still varied. In the literature, BP is related to different scopes and contexts, and is therefore subject to a variety of circumstantial definitions. For example, Graupner et al. (2009) present a definition from Wikipedia that defines BP as “the most efficient (least

amount of effort) and effective (best results) way of accomplishing a task, based on repeata-ble procedures that have proven themselves over time for large numbers of people”. On the

other hand, Szulanski (1996) focuses on presenting the BP as an ideal example of how best to execute a particular task. He explained the importance of viewing a BP, not as a law dictating how things should be done, but rather as an ideal example showing how to do it (Szulanski, 1996). Camp (1989) defines BPs as practices that will lead to the superior performance of a

(35)

35

company. These definitions focus on more than just the content of the BP, as they also implic-itly guarantee that implementing the BP will lead to improvement in performance. Therefore, from the organizational perspective, BP is a good practice since it enhances organizational performance. However, this does not necessarily mean that BP is a good practice for other stakeholders, for example, the customer.

Despite the importance of BPs for KM within organizations, the conceptual understanding of BP is not without confusion. Fragidis and Tarabanis (2006) argued that the term “best” in BP is not used in a strictly literal manner. Rather than being considered superlative, BPs are sometimes only understood as promising approaches and activities that organizations may consider as useful tools and experimental practices (Fragidis and Tarabanis, 2006). Jarrar and Zairi (2000) fashioned their definition and adopted the Chevron approach by dividing BP into three levels: a good idea, a good practice and a proven BP. The American Productivity and Quality Centre also noted that there can be no all-encompassing BP, because the concept "best" is not universally agreed upon, and each practice has to be adapted to suit distinctive organizational contexts (American Productivity and Quality Centre, 1997). Figure 5 presents the steps of BP that they have proposed. Furthermore, the benefits of BPs are not limited to the sharing of superior practices to accomplish a particular task, but include the opportunity of learning from experiences and the mistakes and failures of others. Therefore, many re-searchers recommend the use of terms such as “good”, “identified”, “smart” and “recom-mended” practices instead of “best” practices (Vesely, 2011).

Figure 5. Four steps of BPs in an organization database, adapted from (American Produc-tivity and Quality Centre, 1997)

A BP is intended to capitalize on an organization’s internal knowledge, for example, by tak-ing advantage of previous practices and experiences, in order to define the best ways to solve a problem. As its core, BP is intended to reuse the best ways to solve a problem or handle an issue. It is all about gaining the benefit from previous experience to define possible ways to conduct activities and solve problems (Axelsson et al. 2011).

In this thesis, a BP is a way of improving a practical example, approach, process, practice, technique or rule for successful implementation of a particular task, where this practice has been applied and, therefore, it is intended to replace an existing practice and to be followed. In this thesis, “practice” in the term ‘best practice’ refers to the repeated performance of a particular task. More precisely, practices are actions repeated over time and space, and em-bedded in a particular context, and are socially developed through learning and training (Ad-ler and Pouliot, 2011). Ad(Ad-ler and Pouliot (2011) present five characteristics of a practice: 1) practice is the process of doing something; 2) practices are actions repeated over time and

(36)

36

space (i.e., they are is patterned); 3) practices are actions that can be done correctly or incor-rectly (and can therefore be validated by an individual or group); 4) practices rest on back-ground knowledge; and 5) practices integrate the discursive and the material world, i.e., the written and spoken communications that describe the practice and are woven together with material artifacts that are used when performing the practice.

2.2.2. Cases of BP Implementations

There are several examples of BP implementations described in the literature in different are-as, such as manufacturing, consulting and IT organizations. Consulting organizations, such as Price Waterhouse Coopers, were reported among the first to initiate and develop BP databases to support their business (O’Dell and Grayson, 1998).

General Motors Hughes Electronics used a BP database repository, in which each BP had a brief description and contact information (O’Leary, 1998).

Siemens documented both their own BPs and those of other organizations in a BP database, which was called Siemens Learning Landscape (Boshyk, 1999). Employees and managers at Siemens could find and retrieve BPs via Siemen’s own intranet. The BP documents provided a learning environment that was exclusively dedicated to Siemens’ specific knowledge, via their intranet.

Cheveron reduced the operation costs by more than $2 billion dollar by sharing BPs, and Texas Instruments increased its annual fabrication capacity of 13 fabrication projects by $1.5 billion dollar by sharing their BPs (O’Dell and Grayson, 1998).

Another example is British Aerospace that established a BP database, which stored BP docu-ments, and called it “Best Practice Centre” which ensured synergy across all areas in the or-ganization “The understanding of knowledge management at British Aerospace can be

de-scribed as a best practice approach: ‘To capture and share best or good practices throughout the company’. We understand this concept of best practice in the following way: ‘If you have a practice or process that has worked (i.e., has created something successful) then everybody should be able to use it as well’." Mertins et al. (2003).

The telecom company, Ericsson, introduced a central committee of managers, which evaluat-ed the company’s BPs (Watson, 2007). These managers Ericsson met quarterly to decide which of the practices were best suited to be shared throughout the organization to convert the practices of all departments. This gave Ericsson a competitive advantage in their production processes through a high degree of standardization.

Related to the implementation of BP is the use of different quality standards. Graupner et al. (2009) call these a Best Practices Framework. In the quality management domain, popular quality standards are ISO 9000 (Peach, 2003), Lean (Boyle and Scherrer-Rathje, 2009) and Six Sigma (Pyzdek and Keller, 2009; Schonberger, 2008). They all provide BP processes and guidelines for higher quality and better performance. In the IT enterprise management and IT governance domains, the Control Objectives for Information and Related Technology (CO-BIT) framework is commonly used for assessing the quality of BPs (ISACA, 2012). The In-formation Technology Infrastructure Library (ITIL) is also a popular framework for IT ser-vices that are used as BPs (Hendriks and Carr, 2002). All these frameworks describe how work should be organized between people within a particular context (Graupner et al. 2009). Table 5 shows an example of a BP. This is a fictitious example that helps supervisors to sup-port students at the department of Computer and System Sciences, at Stockholm University, to refine an initial research idea into a feasible, relevant and well formulated research ques-tion.